What is the relationship between individual happiness, the well-being of citizens and the income per capita of countries? There is a lot of debate regarding this question in academic circles since the ultimate aim of economic policy is to maximise the happiness or well-being of citizens (if possible, the well-being of both present and future citizens). In practice, however, decisions tend to be taken based on criteria for the generation of income (or wealth), assuming this is the best way to ensure citizens are as happy as possible. The appropriateness of economic policy decisions therefore depends on whether income and happiness are closely related.
The pioneering work in this field has been by Richard Easterlin.1 His work suggests that an increase in a country’s GDP per capita is not associated with a rise in the average level of happiness of its inhabitants.2 This finding is known as the «Easterlin paradox» and it has generated a large amount of literature attempting to confirm or refute its findings. Generally speaking, studies on the trend in happiness in a single country over time tend to confirm this paradox. Easterlin himself, updating his initial work,3 shows that real income per capita in the US almost doubled from 1973 to 2004 while mean happiness remained the same. This phenomenon is not restricted to the US, however. The same result was found in Japan, one of the developed countries whose real income per capita has increased the most since the Second World War. The findings are also the same in most European countries.
Nevertheless, this does not mean that money is not important. A lot of literature has shown that, at a specific point in time, there is indeed a very close relationship between income and happiness, both compared across countries and also across individuals in the same country. The key to reconciling these findings with the Easterlin paradox lies in the point in time used as reference: happiness is a subjective and relative concept which makes it difficult to measure over time since personal and/or environmental circumstances can change. This is a very important question and we will return to it later. But first we need to specify the exact relationship between income and happiness, across countries and also across individuals with larger and smaller incomes in the same country and at a specific time.
A country comparison reveals a positive relationship between income per capita and average life satisfaction across countries. Specifically, Stevenson and Wolfers4 documented a correlation of 0.79 between both variables in a sample of 155 countries. They also found that this relationship is what is known as log-lineal: an additional EUR 100 has a greater impact on the happiness of a poor country than a rich one, whereas the increase in happiness will be similar if their respective incomes are doubled.
A comparison between rich and poor individuals in the same country and at the same point in time also reveals a positive relationship between income and life satisfaction. Easterlin had already identified this correlation in 1974 and it has since been confirmed by a large number of studies. For instance, Stevenson and Wolfers analysed this relationship with microeconomic data (at an individual level) in the 25 most populated countries in the world. They found a growing trend in the mean life satisfaction of individuals from less to more according to their income percentile.
Some researchers have proposed a modified (or weaker) version of the Easterlin paradox based on the idea of a satiation point as from which happiness no longer increases with income. This hypothesis implies that, as from a certain level of development, a country’s average happiness no longer increases. Comparing individuals, the impact of an increase in income would also become zero as from a certain income threshold. The study by Stevenson and Wolfers analysed this issue in detail, using various databases and different satiation points. They found that a proportional increase in income produces the same impact on a country’s or individual’s happiness, irrespective of their income level. They therefore conclude that there is no evidence for this satiation point. It should be noted, however, that this analysis measures happiness according to life satisfaction. Kahneman and Deaton,5 in a study only on the US, differentiated between life satisfaction and emotional well-being, referring to the emotional quality of an individual’s everyday experience. While the former appears to have a clearly positive correlation with income, emotional well-being peaks around USD 75,000. This result is found for the three measurements of emotional well-being analysed, referring to the intensity and frequency with which individuals experience enjoyment or joy (positive affect), sadness or worry, and stress.
We have carried out a more or less exhaustive review of the studies analysing the relationship between happiness and income. Now we must reconcile the apparently contradictory empirical findings. On the one hand, happiness increases with income across both countries and individuals; on the other, comparisons over time show a country’s average level of happiness remains constant in spite of economic gains. The generally accepted explanation for this discrepancy is an individual subjectively rates their income in comparison with a benchmark group. In other words, happiness does not depend on a person’s absolute income level. It depends on how someone’s income compares with others (social comparison) or with their own past level (habit).6 This necessarily limits the validity of comparisons over time. The idea is relatively self-explanatory: if the social benchmark were the living conditions of 200 years ago, perhaps all of us would enjoy a much higher level of life satisfaction since the average standard of living today is far above that of even the wealthiest people in the past. Although an increase in income over time has a positive impact on happiness, there are other indirect effects operating in the opposite direction. Ultimately, mean happiness remains relatively constant over time, in spite of society’s higher standard of living.
One general conclusion that might be gleaned from the aforementioned studies is that it is logical to use aggregate income measures, such as GDP per capita, to guide economic policy. This is because the relationship between income and happiness is usually quite close. However, we must also acknowledge the limitations of such measurements and the need to take other factors into account that also determine our well-being and happiness.
Judit Montoriol Garriga
Macroeconomics Unit, Strategic Planning and Research Department, CaixaBank
1. See Easterlin, R. A. «Does Economic Growth Improve the Human Lot? Some Empirical Evidence», in David, P. A. and Reder, M. W. (eds.), «Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz» (New York and London: Academic Press, 1974).
2. Note that happiness is measured subjectively based on responses regarding life satisfaction, such as the Cantril ladder. See the article «Are you happy? Happiness and human beings» in this Dossier for a detailed explanation of the concept of happiness and its measurement.
3. See Easterlin, R. A. (1995), «Will Raising the Incomes of All Increase the Happiness of All», Journal of Economic Behavior & Organization, 27(1), p. 35-47.
4. Stevenson and Wolfers (2013), «Subjective Well-Being and Income: is There Any Evidence of Satiation?», American Economic Review, Papers & Proceedings, 103(3).
5. See Kahneman, D. and Deaton, A. (2010), «High income improves evaluation of life but not emotional well-being», PNAS, vol. 107.
6. See Clark, A. et al. (2008), «Relative Income, Happiness and Utility: An Explanation for the Easterlin Paradox and Other Puzzles», Journal of Economic Literature, vol. 46 (1).